How multi-armed bandits can help Starbucks send personalized offers to its customers.
How often do you wait for a coupon offer to buy a product? And after a few uninteresting deals, did you feel like giving up using that service?
Coupon systems have been widely used to enhance customers’ engagement in digital-based platforms. By offering users a challenge and a corresponding reward, companies’ services become not only more attractive, but most importantly it can lead users to become frequent customers, thus enhancing a brand’s impact on its customers. However, knowing which coupon to provide can be a rather complex task since each customer profile responds differently to each offer, and frequently offering them bad deals might drag them away from your business.
To overcome this problem, machine learning techniques can be used to build data-driven customer profiles and develop better coupon recommendations. For that matter, this article shows how K-Means clustering combined with Multi-armed bandits can be used in the Starbucks Mobile Rewards App to build a coupon recommender system.
This article is the result of a capstone project submitted to Udacity’s Machine Learning Engineer Nanodegree, and its source code can be found in this repository. Please refer to the thorough report on the repository for more technical informations such as parameters sweep and dataset preprocessing.
So without any further ado, let’s get down to business!